Available online at www.sciencedirect.

com

ScienceDirect
Procedia - Social and Behavioral Sciences 111 (2014) 790 – 799

EWGT2013 – 16th Meeting of the EURO Working Group on Transportation

Air Transport Performance and Efficiency:

MCDA vs. DEA Approaches
Maria E. Baltazar a,*, João Jardim a, Pedro Alves a, Jorge Silva a
a
LAETA/UBI-AeroG, Faculty of Engineering Aerospace Sciences Department,Calçada Fonte do Lameiro, 6200-358 Covilhã, Portugal

Abstract

The general aim of this work is the development of airports performance and efficiency predictive models using robust but
flexible methodologies, and incorporating traditional indicators as well as new constraints. Particularly it shows and compares
the efficiency evolution of the same airport along several years, under several constraints, based on two multidimensional
tools: Multi Criteria Decision Analysis (MCDA, by Measuring Attractiveness by a Categorical Based Evaluation Technique -
MACBETH) and Data Envelopment Analysis (DEA).The results evidence how MACBETH (MCDA) approach seems to be a
very promising one when compared with those (DEA based) traditionally in use.

The Authors.
© 2013 The Authors.Published
Publishedby
byElsevier
ElsevierLtd.
Ltd.
Selection and/or
Selection and/or peer-review
peer-reviewunder
underresponsibility
responsibilityofofScientific
ScientificCommittee
Committee.

Keywords: Airports; Performance; Efficiency; MCDA; DEA.

1. Introduction

The airport business has changed quickly over the last decades since it has been a consistent growing segment
inside travel and transportation industries. The annual growth of global aviation industry has sustained rates of
five to six percent (Graham, 2003). More than 5 billion passengers passed through the world’s airports in 2010
(ACI, 2010). However, due to economic downturn demand for air transport slowed in recent years (Fodness and
Murray, 2005). The jet fuel prices and credit crisis has also a negative impact on consumers and consequently in
the number of air travelers. However, new business models adopted by airlines allowed some growth return in the
last years, as the case of low-cost carriers (LCC), being a major proportion of the business volume generated by

* Corresponding author. Tel.: +351965002238; fax: +351275329768.

E-mail address: mmila@ubi.pt

1877-0428 © 2013 The Authors. Published by Elsevier Ltd.

Selection and/or peer-review under responsibility of Scientific Committee
doi:10.1016/j.sbspro.2014.01.113
 Maria E. Baltazar et al. / Procedia - Social and Behavioral Sciences 111 (2014) 790 – 799 791

airports. Off course, mostly this is a political decision and local governments to “attract” the LCC, especially in
small airports, strongly subsidize for example the landing fees.
Airports have become not only nodes of new intermodal transport systems for people and goods, but also new
cities, in a worldwide scale competition (Marques and Galves, 2009). Another important issue, as presented by
(Oum et al., 2003) is the worldwide liberalization of the airline industry. It has increased the demand for more
efficient and faster processing of aircraft, passengers, cargo and baggage. Airport managers are being confronted
with new challenges every day in an era of growing commercial pressures. Thus, it is important for airports to
provide services in the most efficient ways. This work shows the efficiency evaluation of three airport along
several years based on two multidimensional tools: Multi Criteria Decision Analysis (MCDA, particularly
Measuring Attractiveness by a Categorical Based Evaluation Technique - MACBETH) and Data Envelopment
Analysis (DEA): Furthermore it compares the obtained results evidencing the pros and cons of each tool,
searching for the best conditions to apply one or other within airport management decision processes.

2. Scope and purpose

The last years revealed a growing interest in measuring the economic and operational performance of airports
with benchmarking studies inside and outside the airport sector. Airport managers have facing increasingly
requests from government agencies which have sought airport benchmarking as an aid to form or adjust
regulations, and to create legislation (Morrison, 2009).
ACI (2006) describes benchmarking as an important part of an airport’s strategic planning process. It is a
statistical and accounting process that is used to monitor and compare airport economic, operational and service
performance. The airport’s strategic objectives are assessed in order to measure the performance of its functions,
and the best practices for possible incorporation into the organization’s procedures are identified, to increase
efficiency, quality and customer satisfaction.

2.1. Description and Interest

International airports are complex and dynamic organizations providing a challenge in establishing an
appropriate performance measure system. There are many interacting issues that make complex the development
of performance measure systems (airlines, passengers, handling agents, etc.); therefore it is a critical management
activity. The optimization of operational performance is becoming increasingly important to all stakeholders
along the air transport infrastructure. They can be airports or air navigation service providers desiring to improve
their performance in order with strategic business objectives, whilst their customers wish to be assured that
services are being delivered in an efficient and effective way to meet their requirements (Humphreys and Francis,
2002).

2.2. Methodologies to Evaluate Airport Performance and Efficiency

There are two main research lines on airport performance: the productivity evaluation approach and the
efficiency evaluation approach; the difference lies in a concept of maximum attainable outputs. Whereas
productivity considers actual infrastructure outputs, efficiency does not take into account the maximum potential
output which can be produced with the available inputs. The underlying meanings of these two terms are not
identical despite of being often used as synonyms. Also changes in productivity are due to changes in efficiency,
among other factors (Lai et al., 2010).
792 Maria E. Baltazar et al. / Procedia - Social and Behavioral Sciences 111 (2014) 790 – 799

Lay et al. (2010) referred that previous studies often adopted quantitative methods, relying on numerical and
secondary data, in order to evaluate efficiency and productivity: using Total Factor Productivity (TFP) method in
order to examine the performance of six Australian airports over a 4-year period; analyzing airport quality and
performance, from the airline’s point of view using DEA; comparing the relative performance of Spanish airports
using either a Surface Measure of Overall Performance (SMOP) and a DEA; and applying Stochastic Frontier
Analysis (SFA) to a panel data of world’s major airports to study the effects of ownership forms on airport’s cost
efficiency. Methodologies used in several case studies followed MCDA principles (Lay et al., 2010) too.
Following the study of Braz (2011) we developed our work using both a MCDA tool and a DEA approach.

2.3. Efficiency Indicators

Taking into account each airport characteristics and a set of indicators, managers will be in a key position to
decide how many or which indicators that an airport in particular should follow; over time the set of indicators of
an individual airport will change as new issues arise (ACI, 2012).
Since we used two different approaches in this work, for MCDA we used complex indicators (composed by an
output/input structure) and for DEA we used single ones, as shown in Table 1.

Table 1. Single and Complex Indicators

Single indicator (DEA)

Number of Runways; Aircraft Parking Stands; Passenger Terminal Area;
Input Cargo Terminal Area; Number of Boarding Gates; Number of Check-In
Desks; Number of Baggage Carousels
Output Aircraft Movements, Processed Passengers, Processed Cargo (Ton.)
Complex indicator (MACBETH)
PAX/PAX TA Processed Passengers / Passenger Terminal Area
CARGO/CARGO TA Processed Cargo (ton.) / Cargo Terminal Area
MOVS/STANDS Aircraft Movements / Number of Aircraft Parking Stands
MOVS/RWS Aircraft Movements / Number of Runways
PAX/GATES Number of Processed Passengers / Number of Boarding Gates
PAX/CHK-IN Number of Processed Passengers / Number of Check-In Desks
PAX/PAX TA Number of Movements / Number of Boarding Gates
MOVS/BELTS Number of Movements / Number of Baggage Claim Belts (arrivals)

3. Multicriteria Decision Analysis (MCDA) and Data Envelopment Analysis (DEA)

3.1. Multicriteria Decision Analysis (MCDA)

Multicriteria Decision Analysis (MCDA), or Multi Criteria Decision Making (MCDM), is a decision-making
tool aimed to support decision makers who are faced with numerous and conflicting evaluations. It appeared in
1960 in order to highlight these conflicts and deriving a way to compromise in a transparent process. To improve
the quality of decisions involving multiple criteria numerous MCDA methods have been developed by making
choices more explicit, rational and efficient. The aim is to compare a structured process from different
perspectives, identifying objectives and creating alternatives (Marttunen, 2010).
 Maria E. Baltazar et al. / Procedia - Social and Behavioral Sciences 111 (2014) 790 – 799 793

3.1.1. Measuring Attractiveness by a Categorical Based Evaluation Technique (MACBETH)

MACBETH, the acronym for Measuring Attractiveness through a Category Based Evaluation Technique, is a
decision making evaluation method of options within multiple criteria methodologies. The main distinction
between other Multiple Criteria Decision Analysis (MCDA) methods and MACBETH is that it only needs
qualitative judgments about the difference of attractiveness between two elements at a time in order to generate
numerical scores for the options in each criterion and to weight the criteria. The judgments expressed by the
evaluator enter in the M-MACBETH software so their consistency is automatically verified and suggestions are
offered to solve inconsistencies if they arise. Thus, MACBETH decision aid process involves the construction of
a quantitative evaluation model. A value scale for each criterion and weights for the criteria are constructed from
the evaluator’s semantic judgments. The options value scores are subsequently aggregated additively to calculate
the overall value scores that reflect their attractiveness taking all the criteria into consideration (Gómez et al.,
2007).
MACBETH is a Humanistic, an Interactive, and a Constructive tool (Bana e Costa et al., 2003):

• Humanistic: because helps decision makers pondering, communicating, and discussing their value systems
and preferences;
• Interactive: as this reflection and learning process can best spread through socio-technical facilitation
sustained by straightforward question-answering protocols;
• Constructive: the idea that full-bodied convictions about the kind of decision to make do not (pre-) exist in
the mind of the decision maker, nor in the mind of each of the members of a decision advising group, but
that it is possible to provide them with help to form such convictions and to build robust (shared) preferences
concerning the different possible options to solve the problem.

Before the development of any model, and in order to turn the final result more robust, it is necessary the
larger data collection one may obtain about what is going to be studied; this first step led the decision group to
have a global view about the decisions to be taken.
After the indicators choice the next step is to get data needed to fill the performance table of each indicator -
in our case, each airport data. The next step is to create a decision tree with nodes, that is, a decision model; the
nodes correspond to indicators that are going to be taken into account; each decision maker defines the
attractiveness of each indicator in the tree as presented in Figure 1 for the MOVS/RWS indicator (just as an
example). Macbeth divides the scale of attractiveness between its highest value and 0 in seven verbal values: no
difference, very weak, weak, moderate, strong, very strong and extreme, Bana e Costa et al. (2005); after
considering the attractiveness of each node the decision makers must define the attractiveness difference between
each indicator in the model in order to make them consistent at the end.

Fig.1. MACBETH Attractiveness Table (example)

794 Maria E. Baltazar et al. / Procedia - Social and Behavioral Sciences 111 (2014) 790 – 799

After the introduction of these values for each node it is possible to produce a robustness table still giving the
opportunity to the decision maker to adjust the sensibility of the model. Gómez et al. (2007) describe the basics in
the mathematical foundations of this tool. As stated by Bana e Costa et al. (2004), MACBETH has a complex
formulation.

3.2. Data Envelopment Analysis (DEA)

The mathematical tool called DEA provides analysis of different productivity factors and can help the
decision-making of directing the administrative efforts towards the company weakness with the objective of
increasing its performance. As stated before there exists other methods available in the literature, but this
technique besides it is of common application in such studies has been selected by its objectivity and usefulness
for this work.
DEA is a non-parametric method used to measure a firm/infrastructure performance on whatever is produced,
in a DEA parlance, by a decision-making unit (DMU), which in our case will be the airports. This analysis was
earlier proposed by some authors who described it as a mathematical model that provided a new way of obtaining
empirical estimates of external relationships (Martín and Roman, 2006). This was the origin of a based method
on a multi-criteria approach used to evaluate the performance of different DMUs depending on the
multidimensionality of a variety of inputs and outputs. Since then, numerous DEA applications have been used in
different areas, such as education, health care, banking, armed forces, sports, transport areas, agriculture, retail
sources and electricity suppliers covering the basic aspects of DEA models, notation, formulation and geometric
interpretation (Martín and Roman, 2006). DEA is divided into three basic models: variable returns to scale
(VRS), constant returns to scale (CRS), and additive models.
DEA solves a linear programming model for each DMU; for n DMUs n LPPs are solved, with r+s decision
variables. The model is the base for all other models developed in DEA (Meza et al., 2003). Thus, the
relationship between the goods produced (outputs) and the material spent in its production (inputs) is maximized
by defining the weight of each output / input, and taking into account that efficiency of all DMUs, when using the
weight assigned to the analyzed DMU, cannot be greater than the unit value. In this study we used the input-
oriented CCR model, as stated by Ferreira et al. (2010); the software used for this application was the ISYDS
v.3.0 software (Integrated Decision Support System v.3.0).

3.2.1. Integrated Decision Support System (ISYDS)

For Meza et al. (2005) a fundamental step for the development of any DEA software is the set-up and choice
of the algorithm to solve the LPPs associated with this methodology. The Simplex algorithm is widely used for
solving LPPs, and the Interior Points algorithm is mostly used for large scale LPPs (the EMS package uses this
algorithm for solving DEA LPPs). ISYDS uses Simplex algorithm for solving the DEA LPPs. ISYDS uses an
approach which includes a subroutine to avoid degenerating problems. Degeneration is a common problem in
DEA models due to the typical structure of DEA LPPs. Those models present a large number of redundant
constraints for the inefficient DMUs and also a large number of variables and restrictions.
The structure of DEA models often leads to multiple optimal solutions in the multipliers formulation and to
degenerate problems when the envelopment approach is used. ISYDS uses the multipliers formulation and, in the
case of multiple optimal solutions, shows only the first one reached. We used a unique method for solving the
LPPs. The format of the LPPs is variable in order to include different DEA models and orientation. Internally the
input data must be in the proper format (in a matrix structure as in Figure 2) depending on the used model. Data
ordering process in the referred matrix is the most difficult part in the software implementation.
Figure 2 shows a simple data structure, an example from our study cases, in which it’s necessary to indicate:
at first the DMU, input and output numbers (6 DMU, 7 INPUT and 3 OUTPUT, respectively); then the input
 Maria E. Baltazar et al. / Procedia - Social and Behavioral Sciences 111 (2014) 790 – 799 795

(RUNWAYS, STANDS, PASSENGER TERMINAL AREA, ATC, CHECK-IN, GATES and BELTS) and
output data (PAX, MOVEMENTS and CARGO); and finally values for each DMU (PDL2006, (…), PDL2011).

6 7 3
DMU RUNWAYS STANDS ATPAX ATC CHK-IN GATES BELTS PAX MOVS CARGO
PDL 2006 1 9 13637 2200 14 3 3 909609 12165 8593
PDL 2007 1 9 13637 2200 14 3 3 944904 12604 6679
PDL 2008 1 9 13637 2200 14 3 3 925766 12875 6431
PDL 2009 1 9 13637 2200 14 3 3 899266 13449 6245
PDL 2010 1 14 13637 2200 14 3 3 935207 13115 5995
PDL 2011 1 14 13637 2200 14 3 3 933763 12327 5901

Fig.2. ISYDS Entry Data Format (example)

4. Self-Benchmarking Study

An interesting improvement for benchmarking studies is the possibility of both DEA and MACBETH tools to
compare efficiency values of a given airport over several years, e.g., a self-benchmarking, in which the airport
measures its own performance over time. This feature is particularly interesting when observing the answer given
by the airport when there are/were investments in such infrastructure. Thus, this case study performs specifically
a self-benchmarking analysis on three Iberian airports where, over the last few years, some expansion works were
made. This self-benchmarking analysis uses the annual performance data for each airport separately to compare
DEA and MACBETH approaches thus not comparing airports among themselves.
In order to use the MACBETH tool it was necessary to attribute a weight to each indicator; thus, we ask for
the opinion of 28 (national and international) aeronautical specialists (from research, airports, airlines, regulation,
air traffic control, and industry sectors). The average weights attributed to each of the complex indicators of
Table 1 are those of Table 2.

Table 2. Complex Indicators Weights for MACBETH Study Cases

Indicator Weight
MOVS / STANDS 16,61%
MOVS/ RWS 12,78%
PAX / PAX TA 18,01%
CARGO / CARGO TA 12,93%
PAX / CHK-IN 10,93%
PAX / GATES 10,05%
MOVS / GATES 09,56%
MOVS / BELTS 09,09%
100,0%
796 Maria E. Baltazar et al. / Procedia - Social and Behavioral Sciences 111 (2014) 790 – 799

4.1. Lisbon Airport (LIS)

Lisbon Airport (IATA code LIS, and ICAO code LPPT) is an international airport located 7 km (4.3 mi) north
of Lisbon city centre, the capital of Portugal. The airport is surrounded by urban development, being one of the
few airports in Europe located inside a major city. It is operated by ANA – Aeroportos de Portugal. There were
several expansion works at the airport during last year’s, changing: the Number of Parking Stands (STANDS)
due to the construction of new aprons; the Passenger Terminal Area (PAX TA) due to the addition of the
Terminal 2, increasing the Number of Check-In Desks (CHK-IN) and the Number of Boarding Gates (GATES)
as well as a new pier in Terminal 1; and the Cargo Terminal Area (CARGO TA) since it was rebuilt and
expanded. Data for this airport was acquired for the period 2006-2011 (ANA, 2011). The changes in the airport
infrastructure were taken into account just from the next year to that they really happened, since there was no
monthly division.
MACBETH and DEA tools got the airport efficiency ranking based on a combination of the above mentioned
indicators on Table 1, and its related weights on Table2.

Fig. 3. (a) Comparative Efficiency between MACBETH and DEA for Lisbon Airport ; (b) Balance between MACBETH and DEA
Rankings for Lisbon Airport

Figure 3 (a) shows a comparison between MACBETH and DEA efficiency values. As expected DEA values
are higher than MACBETH ones. The main differences between both tools are for 2010 and 2011. Based on
MACBETH efficiency decreased between 2007 and 2010. It is true that the terminal areas were increased in 2008
and 2010 but the number of produced outputs was not enough to increment the global efficiency. However, in
2011the airport began to achieve better results. Lisbon airport got the best value in both approaches for 2007.
Figure 3 (b) shows a comparison between MACBETH and DEA rankings, that is, a comparison among
ranking positions. The main differences are for 2010, probably due to different accuracy of both tools. Curiously
there is an opposite phenomenon for 2007 and 2011: see as in 2007 Lisbon got the 1st position based on
MACBETH and the 5th on DEA, and as in 2011 these positions were reversed between tools. Lisbon airport got
the 1st position for MACBETH on 2007, and for DEA on 2006, 2008, 2010 and 2011. The less efficient years
were 2010 for MACBETH and 2009 for DEA.

4.2. Ponta Delgada Airport (PDL)

João Paulo II Airport (IATA code PDL, and ICAO code LPPD) is an international airport located on the
island of São Miguel, 2 km (1.2 mi) west of the city centre of Ponta Delgada on the Azores Islands, in Portugal.
In terms of traffic, this airport is the busiest in the Azores and is the fourth largest infrastructure managed by
ANA - Aeroportos de Portugal. There were several expansion works at the airport during last year’s, changing
 Maria E. Baltazar et al. / Procedia - Social and Behavioral Sciences 111 (2014) 790 – 799 797

mainly the Number of Parking Stands (STANDS) due to the construction of new aprons. The airport is a hub for
the Azorian airline SATA Air Açores and SATA International. Data for this airport was acquired for the period
2006-2011 (ANA, 2011).
MACBETH and DEA tools got the airport efficiency ranking based on a combination of the above mentioned
indicators on Table 1, and its related weights on Table2.

Fig. 4. (a) Comparative Efficiency between MACBETH and DEA for Ponta Delgada Airport; (b) Balance between MACBETH and DEA
Rankings for Ponta Delgada Airport

Figure 4 (a) shows a comparison between MACBETH and DEA efficiency values. As expected DEA values
are again higher than MACBETH ones. Ponta Delgada airport got the best value in both approaches in the year
2009. The less efficient year was 2011 for both, MACBETH and DEA. Based on MACBETH efficiency
decreased slightly between 2010 and 2011. It is true that the numeral of stands increased in 2010 but the number
of produced outputs was not enough to increment the global efficiency.
Figure 4 (b) shows a comparison between MACBETH and DEA rankings where the main difference was for
2010, probably due to different accuracy of both tools. Based on both approaches the airport got the 1st position
in 2009, and the 6th (and worst) position in 2011.

4.3. Barcelona Airport (BCN)

Barcelona - El Prat Airport (IATA code BCN, and ICAO code LEBL), is an international airport located
12 km (7.5 mi) southwest of the city centre of Barcelona, Catalonia, Spain. It is operated by AENA Aeropuertos,
being and important hub in this region. Data for this airport was acquired for the period 2006-2011 (AENA,
2011). There were several expansion works at the airport during last year’s due to the construction of the new
Terminal 1. MACBETH and DEA tools get the airport efficiency ranking based on a combination of the above
mentioned indicators on Table 1, and its related weights on Table2.
798 Maria E. Baltazar et al. / Procedia - Social and Behavioral Sciences 111 (2014) 790 – 799

Fig.5. (a) Comparative Efficiency between MACBETH and DEA for Barcelona Airport; (b) Balance between MACBETH and DEA
Rankings for Barcelona Airport

Figure 5 (a) shows a comparison between MACBETH and DEA efficiency values. Once again DEA values
are again higher than MACBETH ones. Barcelona airport got the best position in both approaches for 2007;
based on MACBETH the less efficient year was 2010, and on DEA was 2009. The major differences in both
approaches are for 2010 and 2011. Based on MACBETH efficiency decreased between 2007 and 2010. It is true
that there was the construction of the new Terminal 1, but surely the produced outputs were not enough to sustain
the global efficiency.
Figure 5 (b) shows a comparison between MACBETH and DEA rankings where the main differences were
for 2010 (1st on DEA and 6th on MACBETH) and 2011 (1st on DEA and 5th on MACBRTH) probably due to
different accuracy between tools. The airport got the 1st position based on both approaches in 2007, and the less
efficient years were 2010 for MACBETH and 2009 for DEA.

5. Discussion and conclusion

The general aim of this work is the development of airports performance and efficiency predictive models
using robust but flexible methodologies, and incorporating traditional indicators as well as new constraints. As
mentioned, airports are complex infrastructures, located in the middle of a chain of agents, and to promote the
performance of an airport itself also it is necessary to promote that chain as a whole. To achieve such a goal it is
necessary to understand the added value of the airports in particular too, so the choice of the indicators (simple or
complex) to construct the rankings to benchmark the infrastructure must be very accurate. There are several sets
of indicators as well as several benchmarking techniques; however, airport stakeholders need simultaneously
robust and flexible tools, mainly because air transportation acts inside an interactive and iterative world where
changes are sudden and quick.
Our study cases were based on three Iberian airports, two in Portugal (Lisbon and Ponta Delgada) and one in
Spain (Barcelona). All the infrastructures were submitted to several improvements during the recent past. We
chose a set of indicators to perform a self-benchmarking analysis between 2006 and 2011. We used, and
compared, the results of two multidimensional tools: a MCDA/MACBETH one and DEA. The preliminary
results evidenced how MACBETH approach seems to be a very promising one when compared with those (DEA
based) traditionally in use. Mainly because not only MACBETH seems to be more accurate than DEA but also it
can be applied easily in managerial practice involving in the process the stakeholders.
However, improvements / developments must be done in this area mainly those focused in the following
items:
• To search for the most significant sets of indicators to airports managers, to evaluate emerging situations
and/or sudden natural phenomenon that can (also and really) affect the infrastructures performance;
 Maria E. Baltazar et al. / Procedia - Social and Behavioral Sciences 111 (2014) 790 – 799 799

• To search for the best robust and flexible multidimensional tools, that can be used by airport managers in an
user-friendly way;
• To improve the self-benchmarking process, which seemed to deserve a special interest from the majority of
our specialist and all the stakeholders contacted along the work;
• To extend the evaluation of airports performance to economic and hinterland components too, as the airport
itself is only one more element in an integrated chain of multi-actors that needs to be promoted as a whole.

Acknowledgments

To all contacted entities and airport authorities, for their important support and information. To FCT (Foundation
for Science and Technology) - for the MSc grant of João Jardim under AIRDEV (Business Models for Airport
Development and Management) Project, in partnership with MIT-Portugal Program (Reference: MIT-Pt-
TSAAs/0046/2008).

References
ACI – Airport Council International. (2006). Airport Benchmarking to Maximize Efficiency. Genève: ACI World Headquarters.
ACI – Airport Council International. (2010). Annual Report 2010. Genève: ACI World Headquarters.
ACI – Airport Council International. (2012). Guide to Airport Performance Measures. Genève: ACI World Headquarters.
AENA – Aeropuertos de España. (2011). Annual Traffic Reports, 2006 – 2011. Madrid: AENA.
ANA - Aeroportos de Portugal, S.A.. (2011). Annual Traffic Reports, 2006 - 2011, Lisboa: ANA.
Bana e Costa, C., de Corte, J.M., & Vansnick, J.C. (2003). Macbeth, Working paper 03.56. ISBN 0 7530 1520 X. London School of
Economics.
Bana e Costa, C., de Corte, J.M., & Vansnick, J.C. (2004) On the Mathematical Foundations of Macbeth, The London School of Economics
and Political Science.
Bana e Costa, C., de Corte, J.-M., Vansnick, J.-C., Costa, J., Chagas, M., Corrêa, É., João, I., Lopes, F., Lourenço, J., & R. Sánchez-López
(2005). MacBeth User´s Guide, Available: http://www.m-macbeth.com/en/m-home.html. Last access: 30th July, 2013.
Braz, J.,. (2011). O MACBETH como Ferramenta MCDA para o Benchmarking de Aeroportos. MSc Thesis: in Aeronautical Engineering.
Covilhã: University of Beira Interior.
Ferreira, E.Z., Junior, H.V., & Correia, R.C. (2010). Worldwide Efficiency Evaluation of Airports: The use of DEA Methodology. Brazil:
Aeronautics Institute of Technology.
Fodness, D., & Murray B. (2005). Passengers’ expectations of airport service quality, Journal of Services Marketing
Gómez, C., Ladevesa J., Prieto R., Gibert, K.& Valls A.( 2007). Use and evaluation of M-MACBETH.
Graham, A. (2003). Managing Airports; An International Perspective. Oxford: Elsevier.
Humphreys, I. & Francis, G. (2002). Performance measurement: a review of airports. International Journal of Transport Management, 1, 79-
85.
Lai, P., Potter. A & Beynon, M. (2010). The development of benchmarking techniques in airport performance evaluation research. 14th Air
Transport Research Society Conference. Porto: Air Transport Research Society.
Marques, A. & Galves, M. (2009). Application of the Multicriteria Decision Aid (MCDA) to the Industrial Airport Project of Viracopos.
Brazil: Universidade Estadual de Campinas, Faculdade de Engenharia Civil, Arquitetura e Urbanismo.
Martín, J. & Román C. (2006). A Benchmarking Analysis of Spanish Commercial Airports. A comparison Between SMOP and DEA
Ranking Methods. Heidelberg: Springer Science - Springer-Verlag GmbH.
Marttunen, M.(2010). Description of Multi-Criteria Decision Analysis (MCDA). Helsinki: Finnish Environment Institute.
Meza, L., Neto, L., Soares de Mello, J., Gomes, E., & Coelho, P. (2005). Free software for Decision Analysis. A software package for Data
Envelopment models, ICEIS 2005 Proceedings of the 7th International Conference on Enterprise Information Systems, pp. 207-212. Miami:
Florida International University.
Meza, L., Neto, L., Soares de Mello, J., Gomes, E., & Coelho P. (2003). SIAD – Sistema Integrado de Apoio à Decisão: Uma Implementação
Computacional de Modelos de Análise de Envoltória de Dados, VI Simpósio de Pesquisa Operacional Da Marinha, VII Simpósio de
Logística da Marinha. Rio de Janeiro: Escola de Guerra Naval - Urca - Rj
Morrison, W. (2009). Understanding the complexities and challenges of airport performance benchmarking. Airport Management, 3, No.2,
145-158.
Oum, T., Yu, C. & Fu, X. (2003); A comparative analysis of productivity performance of the world’s major airports: summary report of the
ATRS global report benchmarking research report 2002. Journal of Air Transport Management, 9(5):13.